Edgio Inc (EGIO) 2017 Q4 法說會逐字稿

Good day, ladies and gentlemen. Welcome to the Limelight Networks 2017 Fourth Quarter Financial Results Conference Call. (Operator Instructions) Please note, this event is being recorded.

I will now turn the call over to Dan Boncel, Limelight's Chief Accounting Officer. Please go ahead.

Good afternoon, and thank you for joining the Limelight Networks Fourth Quarter 2017 Financial Results Conference Call. This call is being recorded on February 7, 2018, and will be archived on our website for approximately 10 days.

Let me start by quickly covering the safe harbor. We'd like to remind everyone that we'll be making forward-looking statements on this call. Forward-looking statements are all statements that are not strictly statements of historical fact, such as our outlook for 2018 and beyond, our priorities, our pending litigation, our expectations and our operational plans, business strategies and feature functionality announcements. Actual results could differ materially from those contemplated by our forward-looking statements, and reported results should not be considered as an indication for future performance.

For more information, please refer to the risk factors discussed in our periodic filings, including our most recent annual report on Form 10-K. The forward-looking statements on this call are based on information available to us as of today's date, and we disclaim any obligation to update any forward-looking statements, except as required by law.

Joining me on the call today are Bob Lento, our Chief Executive Officer; and Sajid Malhotra, our Chief Financial Officer. We will be available during the Q&A session at the end of the prepared remarks from Bob and Sajid.

I would now like to turn the call over to Bob Lento.

Thanks, Dan, and good afternoon. Today, we announced our fourth quarter and full year results. This was an excellent quarter, capping a year of hard work that resulted in solid momentum in our business and improving financial and operating performance.

Revenue was up 10% for the fourth quarter and was our highest quarterly revenue ever. GAAP gross margin was up 280 basis points over the year-ago quarter and was second only to our record high performance last quarter. Our non-GAAP earnings and adjusted EBITDA set new records this quarter as non-GAAP earnings increased 119% year-over-year and our adjusted EBITDA was up 15% from the year-ago quarter. We are extremely proud of these quarterly results, which were a strong finish to an exceptional year.

For the full year 2017, revenue was up 10% year-over-year and well above our guidance at the beginning of the year and within the range we reiterated last quarter. I'm especially proud of our double-digit revenue growth for the year, which was an important goal going into 2017. Our 2017 gross margin improved by 520 basis points over 2016, and our adjusted EBITDA for 2017 was up 35% from the prior year amount.

These excellent results are an indication that our customer-focused strategy is working. We have momentum in our business and a solid foundation for future financial performance, which we believe positions us well to achieve our financial goals in 2018 and beyond.

In 2017, we made meaningful progress on multiple priorities. Customer satisfaction continued to improve as evidenced by the solid 14-point increase from our 2016 Net Promoter Score. We have earned a higher NPS score every year since our first survey in 2013, totaling an impressive 84-point improvement.

This improved customer satisfaction contributed to record traffic levels in 2017, which were up almost 20% over 2016, while incident tickets were down by a little more than 20%. We are pleased that customers continue to trust us with more traffic as the quality and performance of our network continue to improve.

We focused on geographic expansion in markets important to our customers as part of our customer satisfaction efforts in 2017. We expanded our presence in India, added capacity in Canada, Brazil and several locations in Europe and North America. We also launched expansion projects in the Middle East, Southeast Asia and South Africa, which we expect to finish this quarter.

Carrying this momentum into 2018, we also recently extended our reach into the rapidly growing market in China through an agreement with Tencent Cloud. This agreement allows our customers to access the Chinese market and Tencent customers to access global consumers through our network. We believe these geographic expansion efforts will serve our customers well and provide an important opportunity for additional growth.

We also made numerous and important software enhancements throughout 2017 to drive significant capacity expansion across our network. Our capacity models indicate that since the beginning of 2017, our network capacity has doubled, with over 60% of this increase coming from software enhancements.

These successful software enhancements included the rollout of a new next-generation operating system in the fourth quarter, which has had a positive impact on our network capacity and performance. This new operating system is providing faster throughput for software downloads and higher quality through lower rebuffer rates for video delivery.

One major U.S. video customer has seen its rebuffer performance improve by 19% for high-definition content and 24% for standard-definition content since we deployed the new operating system. Another major global video customer that uses a multi-CDN workflow has been rewarding significantly more traffic to Limelight due to our improved performance. We're excited about the positive impact of this significant development effort.

We began exploring potential opportunities in Edge services in the second half of 2017, as we believe that our large-scale, private global network makes us a good alternative for customers looking to deploy applications at the Edge where latency and connectivity are mission-critical. We have seen growing interest, onboarded a few customers and continue to build our pipeline. I'm very pleased that we just announced our first material win in this space with Avitas, a GE venture, who will be using our global private network and distributed cloud storage infrastructure to deploy its next-generation automated inspection platform.

After scanning the world for performance and completing a 6-month pilot with us, Avitas chose Limelight for its needs, signing a multi-year contract. We are proud to serve Avitas and its large-scale inspection platform, which is enabling its customers to save money and also drive a level of safety and security that is unprecedented.

Turning now to our ongoing legal dispute with Akamai. As you know, we filed a claim for patent infringement against Akamai in the Eastern District of Virginia back in November of 2015. Akamai filed counterclaims asserting infringement of 5 of its patents.

After a series of challenges by both parties to each patent in the case, the court decided to schedule a 1-week trial solely with respect to the question of Akamai's infringement of 3 of our patents. And the trial is scheduled to begin on April 2, 2018, which is less than 2 months from today. We intend to seek the maximum damages allowable by law and also a permanent injunction against future acts of infringement by Akamai with respect to these patents.

Each of Akamai's 5 patents has undergone a validity challenge before the patent trial and appeal board. This has resulted in the invalidity of 1 patent, while the other 4 are still pending. According to statistics cited by Akamai in the Eastern District of Virginia proceedings, once instituted, 2/3 of these types of challenges litigated to a final decision result in the PTAB invalidating the challenged claims. The PTAB will render its validity decisions on Akamai's remaining 4 patents in the coming months, and any patent that survives these challenges will be eligible for a future trial.

Looking ahead at 2018, our long-term strategic priorities will continue to be creating customers for life, growing profitable revenue while generating cash, and improving our position as employer of choice. We are also well-positioned to add significant and innovative features and capabilities and have added this as a key strategic priority.

Customer satisfaction will continue to be at the heart of everything we do in 2018. We will continue to communicate with our customers and focus our efforts in areas that are important to them. The communication between our employees and our customers happens daily, along with our more formal annual customer satisfaction survey and our frequent regional customer advisory forums, serve to keep us well-informed regarding our customers' needs and expectations. We believe this focus on improved customer satisfaction will continue to drive top line growth going forward.

The market for video delivery, both live and on-demand, is growing, and we are focused on becoming a lead provider at scale in this area. We also continue our efforts in Edge services, which we believe has great potential for us, given our unique ability to address customers' needs. We expect to prove our value proposition and exit the year as a key player in this market, which should position us for meaningful growth in 2019 and beyond.

We will continue to expand our geographic footprint in 2018 in locations that are strategic to our customers. We will also build further improvements in our network capacity globally through continued software optimization, which we expect will also drive network efficiencies and better performance. We are encouraged by the capacity increases through software enhancements that we experienced in 2017, and we believe there's still more to come.

Our R&D spend in 2018 will also be focused on driving new feature functionality for our customers, particularly in software downloads, video and Edge services. We have a number of software enhancements and feature functionality releases scheduled for 2018, which we expect to further improve our customers' experience.

We will also focus on our employee satisfaction to maintain our low level of attrition. We will continue to invest in our employees through training and expanded responsibilities, and I expect that employee pride in Limelight will continue to grow.

In summary, this was another great quarter concluding a great year. I'd like to thank our global workforce for all their hard work in making 2017 one of the best years in Limelight's history. I'm very encouraged by our solid momentum on our strategic priorities, and I expect this trend to continue.

We believe that we are well-positioned for continued revenue growth and margin expansion as we remain disciplined in our approach to managing costs and pricing across our customer base. We remain focused on our global goal of delivering the highest quality in the industry, and we believe this will translate into above-market returns.

With that, I'll turn the call over to Sajid to discuss the fourth quarter's financial performance in greater detail and our guidance for 2018.

Thanks, Bob, and good afternoon. I'm happy to report another great quarter as we continue our move to profitable growth in our turnaround. Next milestone is GAAP profitability and positive cash flow, and I will have more on that in our guidance later on.

Revenue was up 10% from the quarter a year ago. It was our fifth consecutive quarter of revenue growth, our second consecutive quarter of double-digit percentage revenue growth, and our highest reported revenue ever. Gross margin at 47.7% increased 280 basis points from the prior year. With operating expenses remaining under control, the combination of these elements resulted in our lowest quarterly GAAP loss since 2012, our best non-GAAP EPS ever and our best quarterly adjusted EBITDA ever.

Revenue in the quarter was $48.2 million, up 10% from Q4 last year. We experienced foreign exchange tailwinds in the quarter of approximately $300,000 or less than 1%, primarily due to the British pound. International customers accounted for 38% of total revenue in Q4 compared to 40% a year ago, and approximately 17% of our fourth quarter revenue was in non-U. S. dollar-denominated currencies.

Our top 20 customers accounted for approximately 69% of total revenue in Q4. Our revenue growth has come from both existing customers as well as new customer wins. Our new customer wins have approximately 3x the amount of annual revenue than the customers leaving us. The positive mix shift has resulted in average monthly revenue per customer to move from $50,000 in the fourth quarter of 2016 to $65,000 in the fourth quarter of 2017. It is our belief that we have one of the highest, if not the highest, average revenue per customer in the industry.

According to the Cisco Visual Networking Index white paper published in June 2017, CDN traffic will carry 71% of all Internet traffic by 2021. This is up from 52% in 2016. We feel we are uniquely positioned to participate in this industry's organic growth. We aim to win market share as we continue to invest in our infrastructure enhancements and add feature functionality that our customers want and need.

Gross margin was 47.7% in the fourth quarter, an increase of 280 basis points over a strong prior year. Cash gross margin was 57.9%, an increase of 190 basis points year-over-year.

Our bandwidth expenses in aggregate dollars increased as we served record traffic in the fourth quarter of 2017. Despite a 12% increase in traffic delivered in the fourth quarter of this year compared to last year, our bandwidth expense was essentially flat as we continue to efficiently manage this and every aspect of our cost line.

[Rack] fees remained consistent with the prior year despite the addition of multiple geographies and the significant increase in overall capacity. Rack fees also continued to decline as a percentage of revenue, contributing to our gross margin improvements.

The increase in gross margin at Limelight has been incredible, and we expect it to continue to make further improvements in the coming years. Customer and product mix as well as the new revenue streams we mentioned around security and Edge services should help this effort for the foreseeable future.

Total GAAP operating expenses were $24.2 million, which is an increase of $1.3 million from the fourth quarter of 2016. G&A expense increased by $700,000. Sales and marketing expense increased by $800,000. These increases were partially offset by a decrease of $100,000 in R&D expenses year-over-year. With our strong performance, variable compensation expense is up and is captured in operating expenses across all departments. It is included in other current liabilities on our balance sheet.

We had net interest income of $90,000 in the fourth quarter of 2017 compared to $40,000 last year. Other income was $200,000 in the fourth quarter of 2017 compared to $600,000 of expense last year. This change was primarily due to foreign currency fluctuations.

GAAP net loss was $0.01 per share in the fourth quarter of 2017, and we achieved a strong $0.04 in non-GAAP earnings per share. We reported $0.04 GAAP net loss and positive $0.02 per share of non-GAAP earnings in the fourth quarter of 2016. Tax reform did not have a material impact on our financial results, given our net operating loss position and our valuation allowance on that NOL. Adjusted EBITDA was approximately $8.7 million, up 15% from $7.5 million in the fourth quarter of 2016.

Moving to the balance sheet and cash flow. We had cash and marketable securities of $49.4 million at the end of the fourth quarter. The primary drivers of our change in cash were an increase in accounts receivable of $3.7 million and a decrease in accounts payable of $4.4 million. We expect these year-end calendar-related changes to correct as early as the coming quarter.

We also made our 6th of 12 $4.5 million payments from the previously announced settlement with Akamai, and we spent $4.9 million on capital expenditures in the quarter. Capital expenditures totaled $20.7 million for the year, in line with our original guidance.

DSO as of December 2017 was 62 days, 7 days higher than the end of the fourth quarter 2016, as it looks like some customers were managing their cash at the end of the year. We expect DSO to be more in line with our historical averages and be in the mid-50s in the first quarter of 2018.

We maintain our borrowing capacity under our revolving credit agreement at $10 million, although we have not drawn on it to date.

During the fourth quarter, Goldman Sachs, our largest shareholder, sold 50 million shares or roughly half of their position. Earlier today, we filed a shelf registration statement with the SEC to register the remaining shares held by Goldman. Applicable securities laws prevent us from further commenting on the subject until the registration statement becomes effective. As of December 31, we had approximately 111 million shares outstanding.

Total headcount at the end of the quarter was 533, up 23 from the end of last year. The increase is related to additional operational support and sales and marketing personnel.

Moving to our guidance for 2018. We expect revenue to be between $196 million and $200 million. We expect gross margins to increase by more than 100 basis points over 2017, and we expect non-GAAP EPS to be between $0.11 and $0.15. Adjusted EBITDA is expected to be between $32 million and $36 million. Capital expenditures are expected to be between $22 million and $24 million.

Within the year, we expect the four quarters to be more even than in previous years. If that holds, we could be close to breakeven at the operating income line for the four quarters. Our goal is to increase our cash balance every quarter before any buyback, M&A, et cetera. These metrics have eluded us as a public company, but I believe it's time to put these down as a near-term target.

In summary, 2017 has been our finest year across many, many fronts. We have now reported 5 consecutive quarters of revenue growth. We added more revenue in 2017 than we did collectively between 2011 and 2016. We improved gross margin by over 520 basis points. Record revenue, record cash and GAAP gross margin, record non-GAAP net income, record EBITDA, record adjusted EBITDA.

When I look back as where our 3-year analyst estimates were at the beginning of 2017 and compare them to what we achieved this year, it is remarkable. For example, we nearly achieved 2018 analyst estimates from our 2017 performance. On multiple metrics, our guidance for 2018 now is higher than what the estimate said for 2019 a year ago. On profitability measures, we are essentially more than a year ahead of where consensus views were a year ago. We are very proud of this achievement, and we look to build on this momentum moving forward.

As we look ahead, we are increasingly confident about our ability to play a meaningful role in businesses requiring low-latency Edge services. Let me spend a minute on why. The entry into new business typically goes through a buy versus build analysis. The buy decision is a quick start, immediate addition but carries higher operational, integration and balance sheet risks. On the other hand, the build decision is less risky with capital deployment and the ability to review and revise direction, but it is inherently slow. With our Edge services initiatives, which we see as an expansion into an adjacency, we get speed; our global network is ready. At the same time, there is very little risk, almost no incremental capital deployment, no big balance sheet [debt], and we come to the market with a reputation, traction and existing relationships.

Relative to our scale, in my career, I have never come across an opportunity so large at a starting cost so low. Our early wins are with marquee names, as evidenced with the Avitas announcement. Our competitive advantage against even the biggest names is significant as we bring low-latency access with a secure network, high-performance storage, and connectivity at a global scale. And compared to our CDN business, we expect longer-term contracts with healthier margins and lower customer flight risk.

For these reasons, we believe this could be a significant source of new revenue with respectable margins for years to come. At the same time, we will continue to stay focused on growing the core, improving margin, managing operating expenses and achieving profitability. This is an exciting time for all attached to Limelight.

With that, we'll open the call up for your questions. Operator?

(Operator Instructions) The first question comes from Mark Kelleher with D. A. Davidson.

Thanks for taking the questions, and congratulations on some really good execution there. Just a couple of housekeeping first. Were there any 10% customers in the quarter?

Yes, there's one customer. If you recall last quarter, we had 2. This quarter, we have one customer. It's Amazon.

Okay. And were there any -- can you tell us what the pricing dynamics are for content delivery these days? Are you seeing any competitive challenges there?

Nothing out of the ordinary. I mean, historically, we have price compression every year. We try to manage that as best we can, both in terms of passing along the savings that we achieve through cost management in our business and at the same time, dealing with competitive pressures. So I don't really see any big change at a macro level. Obviously, on any one deal, one competitor can get a little crazy, and we see that from time to time. But at a macro level, I think it's been pretty steady.

Okay. And last question. Just can you talk a little bit about that Tencent partnership; what can we -- what should we expect from that? When does it ramp? What does that look like?

So obviously, lots of content being created inside of China that with a large population of Chinese-speaking people outside of China, Tencent wants to help its customers deliver that content to wherever Chinese-speaking people are. And obviously, we want to help them do that through our network. So that's one side of it.

And we have lots of customers that also, at the same time, want to get their content into China. We made a decision years ago not to build out or try to build out in China and instead, to partner. And what we see with Tencent cloud is a very large, mature, high-quality network, much like we have outside of China. And so I think it's going to be very good for both of us.

In terms of ramping, we've actually -- we are actually already using Tencent for at least 1 or 2 -- and keep in mind, we just signed this a few weeks ago -- 1 or 2 of our larger customers are just starting to ramp with us inside of China using the Tencent capabilities and network. And we are in the process of working with them on opportunities for them to use our network. So it's early days. We like what we're seeing in terms of the quality they can provide, and we're actively talking with our customers about that. And we're just starting to talk to them about moving traffic into our network on behalf of their customers for content outside of China.

So we should think of this as a 2019 impact?

Yes, Mark, I think in terms of percentages, if we just wanted to put the marketing spin on it, it's a lot of very large percentage improvements, both this as well as what we see from the other big announcement from the quarter along Avitas. But it's factored into our guidance for this year. I think it ramps up over the course of the year, being more meaningful in the second half of the year, but this is -- this could be material for us in 2019 and beyond.

The next question comes from Jon Charbonneau with Cowen.

In terms of traffic, how fast did traffic grow year-over-year in the fourth quarter? And any change in terms of the mix between OTT and software downloads? I believe last quarter, you noted an acceleration in OTT. Did that continue in the fourth quarter?

Yes, we're still seeing that more. A greater percentage of our traffic is coming from everything video. And so we're pleased with that because that's where a lot of our R&D dollars are going into. And in fact, we have some pretty amazing feature functionality items that will be coming out within the first half of this year, especially around low latency, which obviously is great for live sports and other types of live activities, gambling and other things, where that really matters.

In terms of traffic, the fourth quarter was about 12% year-over-year. And for the year, it was about 20%. So those are kind of round numbers. So we're pleased -- especially pleased that we're able to hit the revenue that we did in the fourth quarter without having to take -- I mean, 12% is still healthy, but lower than what we did on average for the year and still not only hit, but exceed the expectations that we set both publicly and for ourselves in the fourth quarter.

And Jon, I would just add to that, that if you do the math around traffic growth in the quarter or for the year, and then map that against our revenue, this is probably one of our better performances in terms of price compression. And so whatever you may hear in the marketplace about price and aggressive pricing, et cetera, I think we did a really good job of managing our pricing and the discipline around that and fixing our -- or focusing on the business that attracts quality as opposed to indiscriminate pricing.

The next question comes from Michael Turits with Raymond James.

Just wanted to follow up on that. So what was behind traffic slowing from -- throughout the year? Was it 18%, what, 2 quarters ago, then 12%; so it was 20%, then 18% then 12%. So why was it slowing? And did that have anything to do with Akamai claiming that they're getting more aggressive with their top customers?

Yes, we can't speak to what Akamai is or is not doing. Obviously, listened to their call yesterday and quite frankly, confused by their numbers. But that aside, it's not a slowdown of traffic. We made some decisions in terms of what traffic we would choose to handle in the fourth quarter and chose to be less aggressive in the area of what I would call the bulk mail delivery, the software download kinds of traffic. And at the same time, very, very focused on the video traffic, which, on average, carries a higher ASP than the software download traffic.

And so for us, it was a conscious decision. And the fact that Q4 was a little bit lower than the year on average, I wouldn't read much into that. We're -- this year, we expect the number to be higher than 12% from a traffic growth standpoint. So that -- really, in the Q4, we made some decisions in terms of how we were going to go about meeting our financial expectations, and we were lucky enough to be able to get enough of the video business to be able to reach our financial goals without having to go after the low-price bulk traffic.

And Mike, I would just add to that, that that is evidenced in the revenue acceleration. So if you look at the first half of the year, the first and second quarter versus our third and fourth quarter, our revenue ramped up. So we had less price compression. We stopped going after all of the traffic that's available at some really, really, really -- at pricing that we just don't find attractive. Others do. They're chasing it. It's evident. If somebody's talking about high growth in traffic, but it's not showing up in revenue, and that's the business that they want to pursue, that's good for them. It's not good for us.

That's the confusing part. I mean, we had 12% increase in traffic, as we said, with a 10% increase in revenue. If I read the Akamai notes properly, they were down 3% or 4% in revenue but claimed that traffic was growing in excess of industry rates. So I don't -- there's not too many variables in there to make that happen, but I'm not sure how that works.

What about with the large cloud vendors -- Apple, Facebook, Microsoft, et cetera. You've commented on some of these guys in the past. Can you comment at all in any way on what your traffic was like for those guys?

Yes, we don't do business with all of the names that are in their list of 6. But with the names that we do business, our revenues are up with them as a group.

And the relationships are good. So we've already talked about the fact that Amazon is our 10% customer this last quarter. Apple was in Q3. The fact that they weren't in Q4, that relationship is still healthy. Our traffic is growing with them. Obviously, they had a big operating system release in Q3 that pumped that up a little bit. So we enjoy a healthy relationship with some of those customers. And some of those customers, we don't do any business with and never have for that matter.

The last question, just a clarification, Sajid. I think -- what was your guidance for 2018 in terms of, I think -- was it net income positive? And did you say for each of the quarters? I just want to make sure I was really clear on what you meant.

Obviously, we want to be able to do it for the full year. But if the quarterly revenue is about even, we'd love to be able to do it for the 4 quarters or try and get as close to it as possible.

You're talking about net income now?

Yes.

Okay. So you're shooting for net income positive each quarter?

Correct.

The next question comes from Greg McDowell with JMP Securities.

A lot to be proud of. I want to first focus on the new customer wins having 3x the annual revenue of customers leaving you, and maybe help us think about 2018 and number of active customers and how that all play out. Because I know over the previous years that that customer number has been coming down. But maybe just a little guidance on how we should be thinking about that number moving forward. And then I have some follow-ups.

Yes, so let me make a comment on it, and then I will pass it over to Sajid. So if you remember back in 2016, we cut back on our sales and marketing expense and we started to rebuild that as we went into '17. But the actions you take in '16 affect the results in '17. So '17, we started building back that expense, and we are further increasing our sales force and marketing spend in '18 over '17. So we expect that we will add more customers this year than we did last year. But from a focus standpoint, we've really made the decision to try to focus on the medium to large-sized enterprises. And generally speaking, the customers that have been leaving are the smaller ones, and the customers that we've been going after are more of the marquee names in the industries that we choose to participate in. As we go through 2018, our desire is to close the gap between customers leaving and new customers coming in, still focusing on the largest customers, but continuing to do a better job of reducing churn. 2017 was a record year for us, but so was '16 and '15 and '14 and '13. So we've been getting better at that, and so we'll continue to focus on reducing churn but much more aggressive in terms of the amount of dollars that we're deploying to win new customers. Sajid?

I'd just resonate with that. I think it's one of those statistics that we've been publishing for some time. When I meet with investors or with people like yourself, I remind them that this is not a statistic that is really a key driver of our business one way or the other, but we continue to publish it because people ask for it, because they're used to it, and so you can see the progress or the lack of progress on any one number, et cetera.

Having said that, the idea is to continue to go after customers that want our services and our feature functionality and the quality that we bring. And I think that there is a market out there, and I think we continue to pursue those customers. And some of the customers we've shared, I think we can argue if we had any business serving them in the first place, and they've actually helped our gross margin story. Because we are largely a direct high-touch business. The idea of servicing a customer that has a sub-$500 revenue stream is a very, very expensive proposition for us. So exchanging those customers out and getting customers who are spending in the thousands with us is a better deal for us every time.

Absolutely. And Sajid, one follow-up for you, a two-part financial question. First, I just want to be entirely clear on your commentary on the seasonality of revenue for the year, and whether or not we should see more of a traditional ramp-up quarter-by-quarter or if what you're saying, if we took the high end of guidance of $200 million and divide it by 4, that would be $50 million a quarter. But my fear in doing that is, on a sequential basis from Q4 to Q1, historically, we haven't seen that sort of revenue bump.

So that's part A. I just want to better understand those revenue seasonality dynamics. And part 2 is just further -- if you could just comment further on cash flow and the seasonality of cash flow in light of your comments of what transpired in Q4.

Sure. So listen, if you look at our history, we report a fairly strong Q1 and we've reported a couple of lousy Q3s, without mincing words. Nothing we were proud about. And the revenue stream in Q3 has dipped quite a bit, at least for 2 out of the last 3 years. Last year was much better.

Now if you look at last year's performance, the 4 quarters were closer to each other than they historically had been. And I think that that is closer to reality than what was happening 2 years ago or 3 years ago or 4 years ago. So I don't think you can kind of just go $200 million and divide by 4, but I am suggesting that the variability within the quarters won't be $4 million or $5 million; it should be $1 million or $2 million, something like that, right? So we're not going to go from 45, then all of a sudden show up with 36, then jump back up to 51. We are talking about quarters that are closer to the 50 number, right? It's 47, 48, 51, 52, that kind of stuff, and more even across the 4 quarters. Helpful?

Take that, thank you.

Okay, second is your question around the cash flows. So we watch our cash flows, as you know, very, very closely. And $20 million of CapEx, that's what was built into the model last year. We did $20 million -- just a little over $20 million and a little under $21 million. In addition, the big use for cash is the $4.5 million payment that we made to Akamai every quarter. So there's 4 more payments this year and 2 more next year, and then we are done with those.

That aside, typically, our days receivables outstanding runs around 55 days and our payables runs around $7 million, $8 million at the end of the quarter. And as this quarter came about, we went ahead and paid down our payables. I think some of our vendors that we deal with have been -- we've been pushing them to the limit, and it was their year-end as well. And we thought it would be a good gesture to give them the money on time, a little bit early so that they could close their books well, and we did.

And our accounts payable, as you can see in the reported numbers, was quite a bit lower than our historical track record. That will return back to normal. On the other hand, some of the customers are managing their books as well, and we had our receivable days jump up to over 60 days. And I don't think that that is the norm. So we've already collected on some of the big outstanding balances. We are seeing both of those correct back, and I think that will return to normalcy very, very quickly.

The next question comes from Sameet Sinha with B. Riley FBR.

A couple of questions, actually following up on Greg's question. So talking about the even quarters, I would assume that with some of these new businesses that you have between Edge and Tencent that you'd have -- your year should be more second-half weighted. So if you can talk about that.

Secondly, your top 20 customers, so the revenue growth from the top 20 customers is still pretty good, up like 19% year-over-year. Leaving out the third quarter, it's a pretty substantial number, highest in many quarters. But outside of the top 20, it went down by about 5%. So can you talk about -- or help us think about were these the software download business that you did not -- or you decided to not win? Was that in that category, the -- below the top 20 customers?

And lastly, your implied guidance for '18 assumes about -- an OpEx increase of about $10 million. Can you help us think about where all that will go to?

So the increase in OpEx, let's just start there. We are investing more in people, in sales and marketing. I think I've been pretty clear that the launch of the business around the Edge services does not require much capital infusion. In fact, it is -- for anybody else, whether we were starting from scratch or not from the position we have, it would require large amounts of capital. Think about deploying...

Hundreds of millions.

Data centers all around the world, connecting them with security, backbone, storage, et cetera. So we have that advantage. We don't have to do that. But we are going to invest in people to cover the market, get some expertise from the market, establish some relationships and then assist them with the necessary marketing support that they need. So that's part 1.

I think the rest of it is just general growth in the business. I mean, the overall business is growing. We're trying to make sure that OpEx grows at a much slower rate than the overall revenue growth, and that's just a healthy trend. I don't think that we are a bloated company that has to go ahead and kind of go ahead and resize or anything. We are growing in line with and in a very disciplined way in tandem with the revenue growth.

Your next question is around the growth rate of revenue and traffic in the top 20. Our top 20 has been doing very well. I mean, it has historically been overachieving the growth rate of the company overall, and we like that trend. I mean, there's nothing to hide behind that. These are good customers. The 20 customers today are not necessarily the same 20 that we had last year or the year before, so that list keeps changing. But as it changes, the top 20 continues to grow at a very healthy clip and above company corporate averages, and that's fine. That's what we want. We want that group to do really, really well with us because we invest a lot of time focused on their needs.

And your third question was around -- I'm sorry, Sameet, just remind me?

Yes. So my third question -- actually, Sajid, if you can give me a second with that question about the top 20 customers. So the question there was, outside of the top 20, year-over-year decline was about 5%. So the question was, the business that you decided to forgo, was that in the top 20 customers? Or was it outside of the top 20? Because it doesn't make sense if you forgo -- decide to forgo that business from customers outside of the 20.

You know what, the business that we have chosen to become a lot more disciplined around is where there is no -- where there's zero price elasticity. It's somebody who has a bunch of capacity and is willing to give it to anybody at any price, right? So that's what we've forgone. Some of that business is the large customers; they may no longer be in the top 20. They may have dropped off that list. But when we go after a particular type of business, it is not driven by, is this in the top 20 or the next 50 or 80 or in the last 600.

We go after business by what does it really do for our business and utilization of our resources and the margin profile of what we're taking on. So I don't necessarily take into account. I mean, of course, volume matters and larger customers get bigger discounts than smaller customers. But we go after the type of business and where we can discriminate ourselves and differentiate ourselves versus our competition.

And the only thing I would add, I mean, we don't really internally look at like the top 20 versus the rest. But we do look at the top 100 versus the rest. And in our 10% growth, it was pretty even. Our top 100 grew at about 10%, and the rest of the customer base grew at about 10%. So we'd have to follow up on your question about the...

Yes, I need to go ahead and look at those numbers to make sure if that ties together completely.

Right.

Okay. And final question, Sajid, is about the revenue cadence throughout the year. With Tencent and with Edge computing ramping up, I'd assume that second half -- the full year revenue would be more second-half weighted. Is there anything else that we should be considering?

Yes, no, that's a really good question. But when I'm saying that I'm getting to more evenness here, I am expecting uptick here and I am expecting the usual seasonality within the base business. But when it's reported all combined, it should look like very even quarters, or more even than they historically are. So we still have the annual renegotiations with some of our larger customers occurring in the third quarter. We're just able to offset it with other revenue that's coming onboard in plan at about the same time.

Got it. Thank you.

So yes, while it may look more even than it historically is, there's a lot of moving parts underneath the top line.

(Operator Instructions)

Okay, with no other questions in queue, we'd like to thank everyone for their time this afternoon. Please visit the Investor Relations page at limelight.com for upcoming events and presentations. With that, have a good night.

Thanks, everyone.

Thank you.

The conference has now concluded. Thank you for attending today's presentation.